Immunophenotype of Lacrimal Glands in Graves Orbitopathy: Implications for the Pathogenesis of Th1 and Th17 Immunity

Abstract

Background:

Recent studies have reported a wide spectrum of ocular surface injuries in the context of autoimmune reactions in Graves' orbitopathy (GO). Increased expression of inflammatory mediators in tears of GO patients suggests that the lacrimal glands could be a target for immune responses. However, the immunophenotype for GO lacrimal microenvironment is not known. This study aimed to elucidate the pathological changes of GO lacrimal glands.

Methods:

In this case–control study, lacrimal glands were surgically collected from GO patients who underwent orbital decompression surgery and control subjects who underwent other ocular-related surgery. Bulk RNA-sequencing, flow cytometry with dimensional reduction, and immunohistochemical and multiplexed stainings were conducted. Western blotting and multipathway assays were performed in CD34⁺ fibroblasts derived from lacrimal and orbital tissues.

Results:

Increased expression of cytokines and chemokines accompanied by a variety of immune cell infiltrations mainly involving T cells, B cells, and monocytes was found in GO lacrimal glands. An in-depth investigation into T cell subsets revealed interferon (IFN)-γ-producing T helper (Th)1 and interleukin (IL)-17A-producing Th17 cell-dominated autoimmunity in the active GO lacrimal microenvironment. Both fibrosis and adipogenesis were observed in GO lacrimal tissue remodeling. IL-17A, not IFN-γ, stimulated transforming growth factor-β-initiated myofibroblast differentiation as well as 15-deoxy-Δ^12,14-prostaglandin J₂-initiated adipocyte differentiation in CD34⁺ lacrimal fibroblasts (LFs) and orbital fibroblasts (OFs), respectively. IL-17A activated many fibrotic and adipogenic-related signaling pathways in CD34⁺ LFs and OFs. A novel anti-IL-17A monoclonal antibody SHR-1314 could reverse the promoting effect of IL-17A on fibrosis and adipogenesis in CD34⁺ LFs and OFs.

Conclusions:

Our findings provide evidence for the infiltration of different lymphocytes into GO lacrimal microenvironment, where Th1 and Th17 cells characterize the onset of active lacrimal inflammation and contribute to tissue remodeling. These findings may have potential future therapeutic implications regarding the utility of anti-IL-17A therapy, which should be studied in future research.

Introduction

The primary lacrimal glands located within the upper temporal orbit produce the aqueous component of the tear film and play a key role in ocular surface health and visual quality (1). Graves' orbitopathy (GO), the most common extrathyroidal manifestation of Graves' disease, is usually associated with a wide spectrum of ocular surface changes (2,3). Lacrimal gland enlargement and dry eye syndrome in GO patients suggest ocular surface damage resulting from lacrimal gland dysfunction such as hyposecretion that may be caused by the inflammatory process and stimulation of lacrimal thyrotropin receptors by GO autoantibodies (4,5). The significant reduction in corneal surface epithelial cell density in GO patients is probably a secondary lesion due to dry eye syndrome, aggravating ocular surface inflammation (6).

Numerous studies have shown that T helper (Th)1-related cytokines such as interleukin (IL)-12 and tumor necrosis factor (TNF)-α, Th2-related cytokines such as IL-13, and Th17-related cytokines such as IL-1β, IL-6, and IL-17A appeared significantly higher in the tears of GO patients (7 –11). Furthermore, proteomics of GO tears demonstrated the imbalance between inflammatory and protective proteins (12,13). Some proteins, such as cystatin C, α-1-antichymotrypsin, retinal dehydrogenase, and lysozyme C (14,15), displayed different levels in GO tears, which may emerge as promising biomarkers for disease activity and ocular surface disease in GO. Hence, it is of great interest to acquire precise information on the immunopathological changes in lymphocytic distribution and function within the lacrimal glands in GO.

In this study, our aim was to investigate the immunophenotype for GO lacrimal glands and explore for interferon (IFN)-γ-producing Th1 and IL-17A-producing Th17 cell-dominated inflammation in the active GO lacrimal microenvironment. Understanding the altered regulation of autoimmunity in lacrimal glands will inform the potential future treatment for GO.

Materials and Methods

Study subjects

The study design is a case–control study. We screened 76 consecutive patients with GO, who underwent orbital decompression surgery at the Department of Ophthalmology, Shanghai Ninth People's Hospital, Shanghai JiaoTong University School of Medicine during the period from June 2020 to June 2021. Patients were eligible if they were 18 years or older, with active or stable GO, where the clinical activity score was evaluated by the same trained rater. The exclusion criteria were as follows: (i) autoimmune diseases, such as Sjögren's syndrome, asthma, and diabetes; (ii) history of exposure to chemical volatilization or strong light in their working environment; (iii) wearing contact lenses in nearly 15 days; (iv) other ocular surface diseases; or (v) current pregnancy or lactation.

All patients had stopped steroid or immunosuppressive agents for at least three months before surgery. We matched by age and sex 25 control subjects with lacrimal gland prolapse, blepharochalasis, or single eyelid, who underwent reconstructive ocular surgery.

The lacrimal glands and orbital connective tissues were collected intraoperatively. Blood samples were taken at the day before surgery. All participants signed informed consents, and the study was approved by the local institutional review board (SH9H-2019-T192-2) and conducted in accordance with the Declaration of Helsinki.

Statistical analysis

The data are presented as mean ± standard deviation or median with interquartile range. The two-tailed Student's t-test, one-way analysis of variance, nonparametric Kruskal–Wallis test, and chi-square test were used to perform comparison among groups, as appropriate. The SPSS Statistics (version 21.0) (International Business Machines Corporation) was used to analyze the data. A p-value <0.05 was considered statistically significant. A Bonferroni-corrected p-value <0.05 was considered to be a significant functional enrichment in the RNA experiments.

A detailed experimental scheme is provided in Supplementary Materials.

Results

A total of 51 eligible patients (26 active GO patients and 25 stable GO patients) were enrolled. The demographic and clinical characteristics of patients and controls are summarized in Table 1.

Table 1.

Patient and Control Demographics

	Active GO	Stable GO	Control	p
No. of subjects	26	25	25	NA
Sex, M/F	13/13	9/16	12/13	0.597
Age, mean ± SD [range], years	48.92 ± 11.59 [18–63]	41.84 ± 12.54 [23–69]	42.12 ± 11.66 [20–65]	0.061
Smoking history	4/26	3/25	0/25	NA
Thyroid status
Hyperthyroidism	2/26	1/25	0/25	NA
Hypothyroidism	2/26	0/25	0/25	NA
Euthyroidism	22/26	24/25	25/25	NA
Clinical activity score, median (IQR) [range]	3 (3–4) [3–5]	2 (1–2) [1–2]	/	<0.001
Severity grading
Moderate-to-severe	4/26	22/25	/	NA
Very severe	22/26	3/25	/	NA
Thyroid hormone
Free triiodothyronine, median (IQR), 2.3–4.8, pg/mL	3.20 (2.85–3.63)	3.09 (2.83–3.44)	/	0.802
Free thyroxine, median (IQR), 0.62–1.24, ng/dL	0.94 (0.80–1.03)	0.82 (0.75–0.94)	/	0.471
Triiodothyronine, median (IQR), 0.6–1.55, ng/mL	1.04 (0.88–1.12)	0.99 (0.82–1.11)	/	0.793
Thyroxine, median (IQR), 5.42–12.74, μg/dL	8.74 (7.88–10.04)	8.48 (7.59–9.18)	/	0.460
Thyrotropin, median (IQR), 0.56–5.91, μIU/mL	0.74 (0.05–1.77)	1.63 (0.79–3.31)	/	0.539
TRAb, median (IQR), 0–1.75, IU/L	5.82 (2.49–14.85)	1.09 (0.80–2.97)	/	0.031
Thyroid treatment
Antithyroid drugs	8/26	5/25	0/25	NA
Thyroidectomy	0/26	0/25	0/25	NA
Levothyroxine	4/26	4/25	0/25	NA
None	14/26	16/25	25/25	NA

F, female; GO, Graves' orbitopathy; IQR, interquartile range; M, male; SD, standard deviation; TRAb, thyrotropin receptor antibody.

GO lacrimal glands were infiltrated by different immune cell subsets

Our RNA-sequencing data showed that the gene expression of CD14 (monocytes), CD19 (B cells), CD3E (total T cells), CD8A (cytotoxic T cells), PTPRC (total leukomonocytes), KIT (mast cells), CD38 (plasma cells), CD4 (Th cells), and NCAM1 (natural killer cells) was highly elevated in the active group compared with the stable group and normal controls (Fig. 1A). T cells, B cells, and monocytes were the most remarkably increased immune cell subsets in active lacrimal glands (Fig. 1A). Further analysis of cytokine and chemokine gene expression showed that they were more concentrated in active group than the stable and normal groups. TNF, IL-1B, IL-16, PDGF, CCL2, CXCL9/10/11, and IL-6 were increased in active lacrimal glands (Fig. 1B), consistent with previous findings in orbital connective tissues (16). Th1 and Th17 cytokine genes IFNG and IL-17A were elevated in active lacrimal glands, while the Th2 cytokine gene IL-4 was higher in normal lacrimal glands (Fig. 1B).

FIG. 1.

Immune cell infiltration and cytokine and chemokine expression in lacrimal glands from GO patients and control subjects. (A) Gene expression of cell lineage markers in active, stable, and control lacrimal glands. (B) Gene expression of inflammatory cytokines and chemokines in active, stable, and control lacrimal glands. (C) Expression intensities projected on the corresponding TriMAP plots. (D) Distribution of active (purplish red), stable (yellow), and control (blue) lacrimal glands in plots derived by TriMAP. (E) TriMAP plots showing expression patterns for the indicated cell lineage and cytokine markers. (F) Histograms showing expression of the indicated cell lineage and cytokine markers in active (purplish red), stable (yellow), and control (blue) lacrimal glands (RNA sequencing: five active GO, three stable GO, and four controls; TriMAP analysis: three active GO, four stable GO, and three controls). GO, Graves' orbitopathy.

These data suggest that the inflammatory responses were stronger in active lacrimal glands and that both Th1 and Th17 cells would be the mainly involved in T cell subsets in GO lacrimal inflammation.

GO lacrimal inflammation was characterized by IFN-γ-producing Th1 and IL-17A-producing Th17 cells

To confirm the findings at transcription levels, we examined the main immune cell lineages using flow cytometry. Several clusters were formed in the TriMAP maps according to the defined lineage markers and all clusters contained populations found in active, stable, and normal lacrimal glands (Fig. 1C–E). There were two clearly separated clusters of CD3⁺ cells, one uniformly coexpressing CD4, and the other with more variable levels of CD8. IFN-γ was variably expressed in both of these two clusters, while IL-17A was expressed mainly in CD4⁺ cluster and IL-4 was not differentially expressed in all clusters (Fig. 1E). The cluster located in close proximity to the CD4⁺ cells was characterized by a high expression of CD14, indicating monocytes (Fig. 1E). At the right of CD14⁺ cluster was CD19⁺ cells, suggesting that these were B cells (Fig. 1E).

Moreover, our data demonstrated that the expression of CD3, CD4, CD8, CD19, CD14, IFN-γ, and IL-17A was higher in GO lacrimal glands than control ones, especially in the active stage. The expression of IL-4 was much lower and there were no apparent differences among the three groups (Fig. 1F).

In line with the flow cytometry data, a histological study showed that CD3⁺ cells infiltrated GO lacrimal glands compared with normal ones, especially in the active stage (Fig. 2A). In addition, those infiltrative T cells in GO lacrimal glands included both CD4⁺ (Fig. 2B) and CD8⁺ (Fig. 2C) subsets, which significantly gathered focally in the active stage. Furthermore, the expression of Th1 and Th17 cytokines IFN-γ and IL-17A was much higher in active lacrimal glands (Fig. 2D, E), suggesting an early phase of GO lacrimal autoimmunity characterized by Th1 and Th17 cells. This was in agreement with previous findings in orbital connective tissues (17,18).

FIG. 2.

Immunohistochemical staining for lacrimal glands from GO patients and control subjects. The infiltration of CD3⁺ T cells (total T cells) (A), CD4⁺ Th cells (B), CD8⁺ Tc cells (C), and expression of IL-17A (D) and IFN-γ (E) in active, stable, and control lacrimal glands (original magnification, 200 × and 400 × ). Representative data were shown (immunohistochemical staining: five active GO, five stable GO, and five controls). IFN, interferon; IL, interleukin; Th, T helper; Tc, cytotoxic T.

Heterogeneous T cell subpopulations were observed in GO lacrimal glands

We next assessed the spatial distribution of T cell subpopulations in GO and control lacrimal glands by multiplexed immunohistochemistry (IHC). Our results showed the expected patterns of cell staining, with CD4 delineating cells morphologically consistent with Th cells, and CD8 delineating cells morphologically consistent with cytotoxic T cells (Fig. 3A). Tbet, GATA3, retinoic acid receptor-related orphan receptor (ROR)γt, and FOXP3 are key transcriptional factors for Th1, Th2, Th17, and regulatory T (Treg) cells, respectively (19). Their coexpression with CD4 and DAPI highlighted the presence of those T cell subsets (Fig. 3A, B) within GO lacrimal microenvironment, especially in the active stage (Fig. 3A). Both CD4⁺ and CD8⁺ T cells and the Th1, Th2, Th17, and Treg cell subsets were almost invisible in a normal lacrimal microenvironment (Fig. 3A).

FIG. 3.

Spatial distribution of T cell subsets in lacrimal glands from GO patients and control subjects by multiplexed immunohistochemistry assay. (A) Images of active, stable, and control lacrimal glands stained for CD4 (green), CD8 (red), Tbet (powder blue), RORγt (purple), GATA3 (orange), FOXP3 (pink), and nuclei using DAPI (dark blue). (B) The morphology of Th1 (CD4⁺Tbet⁺), Th17 (CD4⁺RORγt⁺), Th2 (CD4⁺GATA3⁺), and Treg (CD4⁺FOXP3⁺) cells, coming from (A). Only two colored markers were shown for each Th cell subset for clarity. (C) Automated image analysis for the quantification of CD4⁺ Th, CD8⁺ Tc, CD4⁺Tbet⁺ Th1, CD4⁺RORγt⁺ Th17, CD4⁺GATA3⁺ Th2, and CD4⁺FOXP3⁺ Treg cells in active, stable, and control lacrimal glands. (D, E) Flow cytometry for the proportions of CD3⁺CD8⁻ IFN-γ-producing T, CD3⁺CD8⁻ IL-17A-producing T, CD3⁺CD8⁻ Tbet-expressing T, and CD3⁺CD8⁻ RORγt-expressing T cells in active, stable, and control lacrimal glands (multiplexed immunohistochemistry: 15 active GO, 18 stable GO, and 15 controls; Flow cytometry: 13 active GO, 15 stable GO, and 12 controls) *p < 0.05; **p < 0.01; ***p < 0.001. ROR, receptor-related orphan receptor; Treg, regulatory T cell.

Quantitative analysis demonstrated that there were a significantly increased number of CD4⁺ (active vs. control, 204.0 [85.0–500.0] vs. 1.0 [0.0–10.0], p < 0.001; stable vs. control, 128.0 [45.0–277.8] vs. 1.0 [0.0–10.0], p < 0.001) and CD8⁺ (active vs. control, 179.0 [114.0–439.0] vs. 20.0 [9.0–31.0], p < 0.001; stable vs. control, 64.5 [32.3–94.5] vs. 20.0 [9.0–31.0], p = 0.033) T cells in both the active and stable lacrimal microenvironments compared with normal control, and the number of CD8⁺ T cells in the active lacrimal microenvironment was higher than that in a stable one (active vs. stable, 179.0 [114.0–439.0] vs. 64.5 [32.3–94.5], p < 0.01) (Fig. 3C).

Meanwhile, the numbers of Th1 (CD4⁺Tbet⁺) (active vs. stable, 37.0 [22.0–93.0] vs. 1.0 [0.0–4.0], p < 0.01; active vs. control, 37.0 [22.0–93.0] vs. 0.0 [0.0–0.0], p < 0.001) and Th17 cells (CD4⁺RORγt⁺) (active vs. stable, 42.0 [23.0–155.0] vs. 6.5 [2.5–17.8], p < 0.01; active vs. control, 42.0 [23.0–155.0] vs. 0.0 [0.0–5.0], p < 0.001) were significantly elevated in the active lacrimal microenvironment compared with those in stable and normal ones. No differences in Th1 and Th17 cells were found between the stable and normal lacrimal microenvironment (Fig. 3C). There was the highest number of Th2 cells (CD4⁺GATA3⁺) (stable vs. active, 18.0 [7.3–55.3] vs. 2.0 [0.0–6.0], p = 0.019; stable vs. control, 18.0 [7.3–55.3] vs. 0.0 [0.0–0.0], p < 0.001) in the stable lacrimal microenvironment among the three groups (Fig. 3C).

In addition, there were more Treg cells (CD4⁺FOXP3⁺) (active vs. control, 6.0 [3.0–15.0] vs. 0.0 [0.0–0.0], p < 0.001; stable vs. control, 5.0 [1.8–19.8] vs. 0.0 [0.0–0.0], p < 0.001) both in active and stable lacrimal microenvironments than in a normal one (Fig. 3C), implying a possible endogenous protective role of immune regulation through the negative feedback loop of Treg cell activation.

We then accessed the frequencies of Th1 and Th17 cells in lacrimal glands by flow cytometry. Significantly increased proportions of CD3⁺CD8⁻IFN-γ-producing T (active vs. stable, 31.7 [21.3–41.1] vs. 3.8 [0.4–20.8], p = 0.015; active vs. control, 31.7 [21.3–41.1] vs. 0.1 [0.0–0.6], p < 0.001) and CD3⁺CD8⁻IL-17A-producing T (active vs. stable, 4.5 [2.9–6.1] vs. 0.7 [0.0–1.4], p < 0.01; active vs. control, 4.5 [2.9–6.1] vs. 0.0 [0.0–0.2], p < 0.001) cells were observed in active lacrimal glands compared with stable and normal ones (Fig. 3D, E), consistent with the TriMAP results.

Similarly, the proportions of CD3⁺CD8⁻Tbet-expressing T (active vs. control, 13.8 [11.3–18.1] vs. 1.0 [0.3–2.6], p < 0.001; stable vs. control, 7.4 [5.3–10.8] vs. 1.0 [0.3–2.6], p < 0.01) and CD3⁺CD8⁻RORγt-expressing T (active vs. control, 36.3 [24.7–53.1] vs. 1.9 [1.4–3.3], p < 0.001; stable vs. control, 17.5 [3.8–29.3] vs. 1.9 [1.4–3.3], p = 0.019) cells were significantly elevated in GO lacrimal glands compared with normal controls (Fig. 3D, E). The proportion of CD3⁺CD8⁻RORγt-expressing T cells (active vs. stable, 36.3 [24.7–53.1] vs. 17.5 [3.8–29.3], p = 0.032) was much higher in active lacrimal glands than stable ones (Fig. 3E).

Fibrosis was a hallmarked pathological change in GO lacrimal glands

Fibrosis, deposition of extracellular matrix (ECM), is an important pathological change of GO orbital connective tissues (20,21). We wondered whether fibrosis also occurred in GO lacrimal glands. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was performed to determine the biological and functional pathways of differentially expressed genes from our RNA-sequencing data. Genes involved in the ECM-receptor interaction pathway of KEGG database are demonstrated in Figure 4A. These ECM-related genes were highly enriched in active lacrimal glands compared with stable and normal controls (Fig. 4A). Since adipogenesis was another vital pathological change of GO orbital connective tissues (20), we examined genes involved in the antiperoxisome proliferator-activated receptor (PPAR) signaling pathway of KEGG database. Likewise, they were mainly elevated in active lacrimal glands (Fig. 4B).

FIG. 4.

Fibrosis and adipogenesis in lacrimal glands from GO patients and control subjects. (A) Genes detected in lacrimal glands and enriched in ECM-receptor interaction pathway of KEGG database. (B) Genes detected in lacrimal glands and enriched in PPAR signaling pathway of KEGG database. (C) The top 20 significantly changed pathways in lacrimal glands between active GO patients and control subjects by KEGG analysis. (D) Eleven significantly elevated genes in ECM-receptor interaction pathway in active GO lacrimal glands. (E) H.E. and specific immunohistochemical stainings for active, stable, and control lacrimal glands (original magnification, 200 × ). Representative data were shown (RNA sequencing: five active GO, three stable GO, and four controls; H.E. and specific immunohistochemical stainings: five active GO, five stable GO, and five controls). ECM, extracellular matrix; H.E., hematoxylin–eosin; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPAR, peroxisome proliferator-activated receptor.

It is not surprising that the fibrotic and adipogenic transcriptome was higher in active stage than in stable stage, implying that there was a sequential process where the fibrotic and adipogenic genes were increased in active disease and gradually faded along with their protein deposition in stable disease marked by tissue remodeling.

Moreover, the ECM-receptor interaction pathway was significantly upregulated (adjusted p < 0.01) in active lacrimal glands compared with normal controls, which was among the top 20 significantly enriched KEGG pathways (Fig. 4C). Within the ECM-receptor interaction pathway, we found 11 differentially expressed genes that were significantly increased, including FN1 and COL1A1 (Fig. 4D). The PPAR signaling pathway was not significantly different in lacrimal glands between GO patients and normal controls.

We also explored lacrimal fibrosis and adipogenesis by the IHC study. Hematoxylin–eosin staining revealed that glandular parenchyma and stroma were infiltrated with a large number of mononuclear cells in active lacrimal glands, indicating robust inflammatory injury. Yet, stable lacrimal glands were destroyed with glandular atrophy, interstitial fibroplasia, and adipocyte proliferation, suggesting vanishing of the inflammation and enhanced tissue remodeling caused by fibrosis and adipogenesis (Fig. 4E). Furthermore, increased ECM deposition with enlarged collagen fiber bundles in GO lacrimal glands was displayed by Alcian blue, Masson, and Sirius red stainings, especially in stable stage (Fig. 4E).

IL-17A promoted fibrosis and adipogenesis while IFN-γ attenuated them in GO fibroblasts

Previous studies showed that circulating CD34⁺ fibrocytes appear to infiltrate GO orbital connective tissues and differentiate into CD34⁺ orbital fibroblasts (OFs) that are the main target cells in GO pathogenesis (22,23). Furthermore, CD34⁺ cells were observed in the fibrous septa of GO lacrimal glands compared with normal ones (Fig. 5A), suggesting that CD34⁺ fibrocytes might possibly infiltrate GO lacrimal glands as well.

FIG. 5.

Effects of IL-17A and IFN-γ on CD34⁺ LFs and OFs derived from GO patients. (A) Infiltration of CD34⁺ cells in active, stable, and control lacrimal glands (original magnification, 200 × and 400 × ). Representative data were shown. (B, F) Western blotting assay showing the differentiation of CD34⁺ LFs and OFs into myofibroblasts induced by TGF-β in the presence or absence of IL-17A or IFN-γ. (C, G) Vimentin staining showing TGF-β-treated CD34⁺ LFs and OFs in the presence or absence of IL-17A or IFN-γ. (D, H) Western blotting assay showing the differentiation of CD34⁺ LFs and OFs into adipocytes induced by 15-D-PGJ₂ in the presence or absence of IL-17A or IFN-γ. (E, I) Oil red O staining showing 15-D-PGJ₂-treated CD34⁺ LFs and OFs in the presence or absence of IL-17A or IFN-γ. Data were combined from at least three independent experiments. 15-D-PGJ₂, 15-deoxy-Δ^12,14-prostaglandin J₂; LF, lacrimal fibroblast; OF, orbital fibroblast; TGF, transforming growth factor.

Our data revealed that transforming growth factor (TGF)-β increased the expression of fibronectin, collagen I, α-smooth muscle actin, and tissue inhibitor of metalloproteinase-I in CD34⁺ lacrimal fibroblasts (LFs) and OFs (Fig. 5B, F). No obvious fibrotic changes were seen if only recombinant human (rh)IL-17A or rhIFN-γ was used to treat CD34⁺ OFs and LFs (Fig. 5B, F). Yet, IL-17A showed a profibrogenic effect in the presence of TGF-β, especially at a high concentration (100 ng/mL) (Fig. 5B, F). In contrast, IFN-γ exerted an inhibitory effect on fibrosis induced by TGF-β in CD34⁺ LFs and OFs (Fig. 5B, F). Our results were further confirmed by vimentin staining in which 100 ng/mL rhIL-17A promoted fibrosis while 5 ng/mL rhIFN-γ attenuated fibrosis in TGF-β-treated CD34⁺ LFs and OFs (Fig. 5C, G).

For adipogenesis induction of CD34⁺ LFs and OFs, the expression of perilipin A, PPAR-γ, antiadipose differentiation-related protein (ADFP), and adiponectin was increased by 15-deoxy-Δ^12,14-prostaglandin J₂ (15-D-PGJ₂) treatment (Fig. 5D, H). Neither rhIL-17A nor rhIFN-γ upregulated those proteins, indicating that IL-17A or IFN-γ alone could not affect the adipogenic differentiation of CD34⁺ LFs and OFs (Fig. 5D, H). On the basis of 15-D-PGJ₂, IL-17A enhanced the expression of perilipin A, PPAR-γ, ADFP, and adiponectin in CD34⁺ LFs and OFs, especially at a high concentration (100 ng/mL) (Fig. 5D, H). However, IFN-γ restrained adipogenesis induced by 15-D-PGJ₂ in CD34⁺ LFs and OFs (Fig. 5D, H).

Likewise, more lipid droplets were observed by oil red O staining in CD34⁺ LFs and OFs in the presence of 15-D-PGJ₂ under the stimulation of 100 ng/mL rhIL-17A. Yet, 5 ng/mL rhIFN-γ decreased lipid droplets in 15-D-PGJ₂-treated CD34⁺ LFs and OFs (Fig. 5E, I). Collectively, our data suggested that IL-17A had both profibrogenic and proadipogenic effects on GO fibroblasts.

IL-17A activated fibrosis and adipogenesis-related pathways in GO fibroblasts

To explore the potential mechanisms of IL-17A on CD34⁺ LFs and OFs, we examined IL-17A-mediated intracellular signaling. The modified protein mass spectrometry of CD34⁺ LFs treated with 100 ng/mL rhIL-17A showed that JAK-STAT, mTOR, and PPAR-γ signaling pathways were activated (Fig. 6A). In CD34⁺ OFs, IL-17A activated JAK-STAT, mTOR, MAPK, adipocytokine, and insulin signaling pathways (Fig. 6B).

FIG. 6.

IL-17A signaling pathways in CD34⁺ LFs and OFs. (A, B) CD34⁺ LFs and OFs were stimulated with 100 ng/mL rhIL-17A for 15 min and were accessed by PTMScan^® Multipathway assay. KEGG database-enriched pathways were presented. (C, E) Reduced fibrosis-related protein expression after SHR-1314 treatment in CD34⁺ LFs and OFs triggered by 100 ng/mL rhIL-17A. (D, F) Reduced adipogenesis-related protein expression after SHR-1314 treatment in CD34⁺ LFs and OFs triggered by 100 ng/mL rhIL-17A. Data were combined from at least three independent experiments. rh, recombinant human.

The effect of SHR-1314, a novel monoclonal antibody of IL-17A

SHR-1314 (vunakizumab; Hengrui Pharma, China) is a humanized IgG1 anti-IL-17A monoclonal antibody composed of two identical heavy chains and two identical light chains. It can selectively bind to IL-17A and block its interaction with receptors.

We investigated its blocking effect in our in vitro GO models. One hundred nanograms per milliliter IL-17A was used as a profibrogenic or proadipogenic agent for TGF-β- or 15-D-PGJ₂-activated CD34⁺ LFs and OFs, respectively. SHR-1314 did not affect TGF-β-induced fibrosis in CD34⁺ LFs and OFs (Fig. 6C, E, and Supplementary Fig. S1A, C), nor did SHR-1314 in 15-D-PGJ₂-induced adipogenesis (Fig. 6D, F and Supplementary Fig. S1B, D). However, SHR-1314 stopped IL-17A-dependent enhancement of fibrogenic and adipogenic processes in CD34⁺ LFs and OFs (Fig. 6C–F). Taken together, our data warranted that the novel monoclonal antibody SHR-1314 could be a promising therapeutic agent for GO.

Discussion

The lacrimal glands are understudied in GO. Although xerostomia and ocular surface damage usually occur in active GO, many studies have confirmed that ocular surface inflammation and dry eye symptoms are quite common not only in the active phase but also in the stable phase (24). The lacrimal gland expressing thyrotropin receptors is histologically identical to the thyroid that is susceptible to immunological damage (2,5), and is likely to be involved in the same autoimmune response in active GO, causing reduced tear secretion and increased inflammatory factors in tear components (7 –11). Therefore, lacrimal glands are expected to be another target organ of GO autoimmune reactions, in addition to orbital connective tissues.

In this study, we demonstrated a complicated GO lacrimal immune microenvironment, mainly consisting of T cells, B cells, and monocytes, characterized by fibrosis and adipogenesis. The increased number of CD8⁺ T cells and monocytes in active lacrimal glands was in conformance with the observation of an early wave of CD8⁺ T cell and macrophage recruitment in orbital connective tissues from GO murine models during disease onset (25). It was previously reported that CD4⁺ Th cells were the most abundant orbital-infiltrating lymphocytes in GO (16,26). We found that the lacrimal CD4⁺ T cells produced higher levels of IFN-γ and IL-17A, especially in the active stage, indicating that both Th1 immunity and Th17 immunity were involved in GO lacrimal inflammation.

Our findings give a clue to the origin of elevated IL-17A levels in GO tears (8 –11). Although gene and protein expressions of IL-4 were not changed in GO lacrimal glands, distinct from orbital connective tissues, the number of GATA3-expressing T cells was significantly increased in stable lacrimal glands, implying a less important role of IL-4 or that other cytokines produced by GATA3⁺ Th2 cells may participate in the pathological process of stable lacrimal glands.

Given the fact that CD34⁺ OFs derived from fibrocytes are the main target cells in GO orbital connective tissues (23), one could speculate that those CD34⁺ fibroblasts may also play a pathogenic role in GO lacrimal glands. As expected, we observed increased generation of CD34⁺ cells in GO lacrimal glands. Our in vitro experiments further confirmed that CD34⁺ LFs differentiated into myofibroblasts or adipocytes, similar to CD34⁺ OFs, under IL-17A stimulation. This could explain lacrimal tissue remodeling formed by fibrosis and adipogenesis attributing to those CD34⁺ LFs. Fibrosis may be due to the activation of JAK-STAT pathway that can promote liver and left atrial fibrosis (27,28) and mTOR pathway that can promote liver and pulmonary fibrosis (29,30) in CD34⁺ LFs and OFs.

The MAPK pathway, which has been proven to mediate GO fibrosis (31,32), was activated in CD34⁺ OFs. Adipogenesis may lie in the activation of PPAR-γ pathway in CD34⁺ LFs, which plays an important role in adipocyte differentiation (33). Yet, in CD34⁺ OFs, the insulin signaling pathway was activated, which we previously reported as enhancing adipocyte differentiation of OFs (34). Altogether, IL-17A contributed to GO fibrosis and adipogenesis via numerous mechanisms, and further confirmatory research is needed.

Shih et al. reported fibrosis in GO Müller's muscles (35). We recently reported an elevated number of Th1 and Th17 cells in GO paranasal sinus mucosa (36). Hence, tissues adjacent to orbital fat may have the same damage mechanisms characterized by inflammation-mediated fibrosis and adipogenesis. More research is needed to determine whether IL-17A or other cytokines participate in those processes.

It is widely accepted that Th17 cells play an important role in human autoimmune diseases, which have been confirmed in disease-associated animal models and clinical practice (19,37). IL-17A blockade has been successful in treating psoriasis and ankylosing spondylitis (38,39). We previously reported that intravenous methylprednisolone therapy significantly reduced the proportion of circulating IL-17A-producing Th17 cells in active, very severe GO patients (40). Mu et al. demonstrated decreased serum IL-17A levels in active, moderate-to-severe GO patients receiving intravenous methylprednisolone either on a weekly or daily regimen at the 12-week visit (41).

However, Siomkajło et al. found no obvious changes of circulating Th17 cells in active, moderate-to-severe GO patients after intravenous methylprednisolone treatment on the same weekly scheme (42). Yet, Siomkajło et al. defined Th17 cells as CD161⁺RORγt⁺ cells and they did not examine the levels of IL-17A. When we defined Th17 cells as CXCR3⁻CCR6⁺ cells, they did not change after intravenous methylprednisolone treatment (40). Owing to the similar proinflammatory, profibrotic, and proadipogenic mechanisms of IL-17A in both the lacrimal glands and orbital connective tissues, targeting Th17 cells may have a therapeutic effect on both tissues.

SHR-1314 is a recombinant humanized monoclonal antibody against IL-17A. At present, 10 clinical trials at different stages have been launched using this drug (e.g., NCT03710681, NCT04840485, NCT05055934). In this study, SHR-1314 appeared to reverse the profibrotic and proadipogenic effects of IL-17A on CD34⁺ LFs and OFs, suggesting a potential future therapeutic benefit.

Our study is subject to several limitations. First, although patients were consecutively recruited, not all of the surgical explants from patients and controls were subject to all of the testing and some of the experiments were performed on small numbers of specimens, and so, there is potential for selection bias. Some barriers to inclusion of all of the samples in all of the analyses included the following: insufficient available tissue volume and the experimental cost. Since the experimental conditions of basic research are relatively controllable, the sample size in each study group of in vitro experiments usually depends on the experimental type and its empirical evidence, with special consideration to the biological repetitions. Different sample sizes of patients and controls were applied for different biological experiments in our study, but the risk of the potential for selection bias still exists.

Second, our study design is exploratory, as we did not have a clear hypothesis and we did not identify an a priori primary outcome. Thus, confirmatory research is needed. Third, the gating strategies of Th17 cells may have influenced the results, which may partially explain the differences in our findings relative to previous studies (40 –42). Since IL-17A may induce fibrosis and adipogenesis both in CD34⁺ OFs and LFs, it would have been preferable and more accurate to focus on the cytokine instead of cell surface marker or transcriptional factor.

In conclusion, our study reveals key features of GO lacrimal immunophenotype in which the occurrence and development of inflammation and tissue remodeling partially depend on IL-17A and IFN-γ axes. Our study suggests T cell-based GO pathogenesis, which will help to inform future research in new therapies for GO.

Footnotes

Authors' Contributions

Research idea: S.F. and H.Z.

Experimental design: S.F., H.Z., and J.S.

Sample collection: S.Z., Y.W., Y.L., and J.S.

Performance of experiments: Y.H., Y.W., S.Z., Y.L., and Y.W.

Data analysis: Y.H., Y.W., and X.L.

Writing and revising: Y.H., Y.W., S.Z., S.F., and H.Z.

Acknowledgment

We thank Mr. Haotian Yu (Majorbio Bio-Pharm Technology) for his help with RNA-sequencing data analysis.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Our study was supported by the National Natural Science Foundation of China (81930024, 81761168037, 81770974, 81800695, 82071003, and 82000879), the National Key R&D Program of China (2018YFC1106100 and 2018YFC1106101), the Research Grant of the Shanghai Science and Technology Committee (20DZ2270800, 17DZ2260100, 19410761100, and 19DZ2331400), the Clinical Research Plan of SHDC (SHDC2020CR3051B), and the Project of Medical Robots (IMR-NPH202002). From the Clinical Joint Research Center of the Institute of Medical Robots, the Shanghai Municipal Education Commission—Gaofeng Clinical Medicine Grant Support (20152228), the Shanghai JiaoTong University Translational Medicine Crossed Research Grant (ZH2018ZDA12, ZH2018QNA07), the Sample Database Project of Shanghai Ninth People's Hospital (YBKB201901), and the Joint Innovation Team for Young Physicians of Shanghai Ninth People's Hospital (QC202002). Innovative Research Team of High-Level Local Universities in Shanghai (SHSMU-ZDCX20210901, SHSMU-ZDCX20210601).

Supplementary Material

Supplementary Materials

Supplementary Figure S1

References

Garg

, Zhang

. 2017. Lacrimal gland development: from signaling interactions to regenerative medicine. Dev Dyn, 246:970–980.

Novaes

, Diniz Grisolia

, Smith

. 2016. Update on thyroid-associated Ophthalmopathy with a special emphasis on the ocular surface. Clin Diabetes Endocrinol, 2:19.

Huang

, Fang

, Li

, Zhou

, Li

, Fan

. 2019. The involvement of T cell pathogenesis in thyroid-associated ophthalmopathy. Eye (Lond), 33:176–182.

Huh

, Kim

, Yoo

, Seo

. 2016. The change of lacrimal gland volume in Korean patients with thyroid-associated ophthalmopathy. Korean J Ophthalmol, 30:319–325.

Eckstein

, Finkenrath

, Heiligenhaus

, Renzing-Köhler

, Esser

, Krüger

, Quadbeck

, Steuhl

, Gieseler

. 2004. Dry eye syndrome in thyroid-associated ophthalmopathy: lacrimal expression of TSH receptor suggests involvement of TSHR-specific autoantibodies. Acta Ophthalmol Scand, 82:291–297.

Villani

, Viola

, Sala

, Salvi

, Mapelli

, Currò

, Vannucchi

, Beck-Peccoz

, Ratiglia

. 2010. Corneal involvement in Graves' orbitopathy: an in vivo confocal study. Invest Ophthalmol Vis Sci, 51:4574–4578.

Kishazi

, Dor

, Eperon

, Oberic

, Turck

, Hamedani

. 2018. Differential profiling of lacrimal cytokines in patients suffering from thyroid-associated orbitopathy. Sci Rep, 8:10792.

Huang

, Luo

, Yang

, Mao

. 2014. Changes of lacrimal gland and tear inflammatory cytokines in thyroid-associated ophthalmopathy. Invest Ophthalmol Vis Sci, 55:4935–4943.

Huang

, Xu

, Song

, Wang

, Yang

. 2012. Inflammatory cytokine profiles in the tears of thyroid-associated ophthalmopathy. Graefes Arch Clin Exp Ophthalmol, 250:619–625.

10.

Ujhelyi

, Gogolak

, Erdei

, Nagy

, Balazs

, Rajnavolgyi

, Berta

, Nagy

. 2012. Graves' orbitopathy results in profound changes in tear composition: a study of plasminogen activator inhibitor-1 and seven cytokines. Thyroid, 22:407–414.

11.

Chen

. 2019. The expression of interleukin-15 and interleukin-17 in tears and orbital tissues of Graves ophthalmopathy patients. J Cell Biochem, 120:6299–6303.

12.

Matheis

, Okrojek

, Grus

, Kahaly

. 2012. Proteomics of tear fluid in thyroid-associated orbitopathy. Thyroid, 22:1039–1045.

13.

Matheis

, Grus

, Breitenfeld

, Knych

, Funke

, Pitz

, Ponto

, Pfeiffer

, Kahaly

. 2015. Proteomics differentiate between thyroid-associated orbitopathy and dry eye syndrome. Invest Ophthalmol Vis Sci, 56:2649–2656.

14.

Kishazi

, Dor

, Eperon

, Oberic

, Hamedani

, Turck

. 2018. Thyroid-associated orbitopathy and tears: a proteomics study. J Proteomics, 170:110–116.

15.

Aass

, Norheim

, Eriksen

, Børnick

, Thorsby

, Pepaj

. 2017. Establishment of a tear protein biomarker panel differentiating between Graves' disease with or without orbitopathy. PLoS One, 12:e0175274.

16.

Fang

, Lu

, Huang

, Zhou

, Fan

. 2021. Mechanisms that underly T cell immunity in Graves' orbitopathy. Front Endocrinol (Lausanne), 12:648732.

17.

Fang

, Huang

, Zhong

, Zhang

, Liu

, Wang

, Gu

, Zhou

, Fan

. 2016. IL-17A promotes RANTES expression, but not IL-16, in orbital fibroblasts via CD40-CD40L combination in thyroid-associated ophthalmopathy. Invest Ophthalmol Vis Sci, 57:6123–6133.

18.

Tang

, Chen

, Mao

, Wan

, Ai

, Yang

, Liu

, Zou

, Lin

, Dan

. 2017. Orbital fibroblasts of Graves' orbitopathy stimulated with proinflammatory cytokines promote B cell survival by secreting BAFF. Mol Cell Endocrinol, 446:1–11.

19.

Dong

. 2021. Cytokine regulation and function in T cells. Annu Rev Immunol, 39:51–76.

20.

Taylor

, Zhang

, Lee

RWJ

, Muller

, Ezra

, Dayan

, Kahaly

, Ludgate

. 2020. New insights into the pathogenesis and nonsurgical management of Graves orbitopathy. Nat Rev Endocrinol, 16:104–116.

21.

Ludgate

. 2020. Fibrosis in dysthyroid eye disease. Eye (Lond), 34:279–284.

22.

Fang

, Huang

, Wang

, Zhang

, Zhong

, Li

, Sun

, Liu

, Wang

, Gu

, Li

, Zhou

, Fan

. 2019. Insights into local orbital immunity: evidence for the involvement of the Th17 cell pathway in thyroid-associated ophthalmopathy. J Clin Endocrinol Metab, 104:1697–1711.

23.

Douglas

, Afifiyan

, Hwang

, Chong

, Haider

, Richards

, Gianoukakis

, Smith

. 2010. Increased generation of fibrocytes in thyroid-associated ophthalmopathy. J Clin Endocrinol Metab, 95:430–438.

24.

Sun

, Zhou

, Fan

. 2021. Ocular surface changes in Graves' ophthalmopathy. Int J Ophthalmol, 14:616–621.

25.

Philipp

, Horstmann

, Hose

, Daser

, Görtz

, Jesenek

, Flögel

, Hansen

, Bechrakis

, Banga

JPS

, Eckstein

, Berchner-Pfannschmidt

. 2022. An early wave of macrophage infiltration intertwined with antigen-specific proinflammatory T cells and browning of adipose tissue characterizes the onset of orbital inflammation in a mouse model of Graves' orbitopathy. Thyroid, 32:283–293.

26.

Wang

, Chen

, Wang

, Guo

, Liu

, Dang

, Hu

, Wu

, Zhang

, Ye

, Shi

. 2021. A novel CD4+ CTL subtype characterized by chemotaxis and inflammation is involved in the pathogenesis of Graves' orbitopathy. Cell Mol Immunol, 18:735–745.

27.

Mair

, Blaas

, Österreicher

, Casanova

, Eferl

. 2011. JAK-STAT signaling in hepatic fibrosis. Front Biosci (Landmark Ed), 16:2794–2811.

28.

Chen

, Surinkaew

, Naud

, Qi

, Gillis

, Shi

, Tardif

, Dobrev

, Nattel

. 2017. JAK-STAT signalling and the atrial fibrillation promoting fibrotic substrate. Cardiovasc Res, 113:310–320.

29.

Lawrence

, Nho

. 2018. The role of the mammalian target of rapamycin (mTOR) in pulmonary fibrosis. Int J Mol Sci, 19:778.

30.

Wang

, Song

, Li

, Wu

, Gu

, Yuan

. 2019. Salvianolic acid A attenuates CCl₄-induced liver fibrosis by regulating the PI3K/AKT/mTOR, Bcl-2/Bax and caspase-3/cleaved caspase-3 signaling pathways. Drug Des Devel Ther, 13:1889–1900.

31.

Fang

, Huang

, Wang

, Zhang

, Luo

, Liu

, Zhong

, Liu

, Li

, Liang

, Miranda

, Gu

, Zhou

, Fan

, Li

. 2016. IL-17A exacerbates fibrosis by promoting the proinflammatory and profibrotic function of orbital fibroblasts in TAO. J Clin Endocrinol Metab, 101:2955–2965.

32.

Fang

, Huang

, Zhong

, Li

, Zhang

, Li

, Sun

, Liu

, Wang

, Zhang

, Xu

, Sun

, Gu

, Li

, Zhou

, Li

, Fan

. 2017. Regulation of orbital fibrosis and adipogenesis by pathogenic Th17 cells in Graves orbitopathy. J Clin Endocrinol Metab, 102:4273–4283.

33.

Kliewer

, Lenhard

, Willson

, Patel

, Morris

, Lehmann

. 1995. A prostaglandin J2 metabolite binds peroxisome proliferator-activated receptor gamma and promotes adipocyte differentiation. Cell, 83:813–819.

34.

Zhao

, Deng

, Gu

, Wang

, Zhou

, Hu

, Chen

, Fan

. 2013. Insulin-like growth factor 1 promotes the proliferation and adipogenesis of orbital adipose-derived stromal cells in thyroid-associated ophthalmopathy. Exp Eye Res, 107:65–73.

35.

Shih

, Liao

, Kuo

, Smith

, Chuang

. 2006. Molecular pathology of Muller's muscle in Graves' ophthalmopathy. J Clin Endocrinol Metab, 91:1159–1167.

36.

, Wu

, Huang

, Fang

, Li

, Sun

, Zhou

. 2021. Immunological features of paranasal sinus mucosa in patients with Graves' orbitopathy. Front Endocrinol (Lausanne), 11:621321.

37.

McGeachy

, Cua

, Gaffen

. 2019. The IL-17 family of cytokines in health and disease. Immunity, 50:892–906.

38.

Burkett

, Kuchroo

. 2016. IL-17 Blockade in psoriasis. Cell, 167:1669.

39.

Dubash

, Bridgewood

, McGonagle

, Marzo-Ortega

. 2019. The advent of IL-17A blockade in ankylosing spondylitis: secukinumab, ixekizumab and beyond. Expert Rev Clin Immunol, 15:123–134.

40.

Fang

, Zhang

, Huang

, Wu

, Lu

, Zhong

, Liu

, Wang

, Li

, Sun

, Gu

, Zhou

, Fan

. 2020. Evidence for associations between Th1/Th17 “Hybrid” phenotype and altered lipometabolism in very severe Graves orbitopathy. J Clin Endocrinol Metab, 105:1851–1867.

41.

, Tang

, Wang

, Lin

, Wang

, Yin

, Shu

, Zhu

, Li

, Zhou

, Zeng

, Chen

. 2020. Comparison of two regimens for patients with thyroid-associated ophthalmopathy receiving intravenous methyl prednisolone: a single center prospective randomized trial. Exp Ther Med, 20:153.

42.

Siomkajło

, Mizera Ł, Szymczak

, Kolačkov

, Grzegrzółka

, Bolanowski

, Daroszewski

. 2021. Effect of systemic steroid therapy in Graves' orbitopathy on regulatory T cells and Th17/Treg ratio. J Endocrinol Invest, 44:2475–2484.